variables.md

Role variables

Role variables are used to configure started instances, cluster topology, vshard bootstrapping, and failover.

Common variables

• cartridge_app_name (string, required): application name;
• cartridge_cluster_cookie (string): cluster cookie for all cluster instances;
• cartridge_not_save_cookie_in_app_config (boolean, default: false): flag indicates that cluster cookie shouldn't be persisted in application configuration file;
• cartridge_remove_temporary_files (boolean, default: false): flag indicates if temporary files should be removed (more details in description of cleanup step API);
• cartridge_ignore_split_brain (boolean, default: false): flag indicates that detected split brain should be ignored on preparation stage;
• cartridge_paths_to_keep_on_cleanup (list-of-strings, default: []): list of paths that are absolute or relative to work/memtx/vinyl/wal directory that should be kept on instance cleanup (config and .tarantool.cookie will be kept independently of this variable); it's possible to use bash patterns, e.g. *.control;
• cartridge_force_leader_control_instance (boolean, default: false): flag indicates that only a leader can be selected as a control instance.

Role scenario configuration

For more details see scenario documentation.

• cartridge_scenario (list-of-strings): list of steps to be launched (see change scenario for more details);
• cartridge_custom_steps_dir (string, default: null): path to directory containing YAML files of custom steps (see change scenario for more details);
• cartridge_custom_steps (list-of-dicts, default: []): list of custom steps (see change scenario for more details);
• cartridge_scenario_name (string, default: default): name of scenario to be launched;
• cartridge_custom_scenarios (dict-of-lists, default: {}): dictionary of custom scenarios.

Application package configuration

• cartridge_package_path (string, optional): path to application package;
• cartridge_enable_tarantool_repo (boolean, default: true): flag indicates if the Tarantool repository should be enabled (for packages with open-source Tarantool dependency).

TGZ specific configuration

• cartridge_multiversion (boolean, default: false): use multiversion approach for TGZ package;

• cartridge_install_tarantool_for_tgz (boolean, default: false): flag indicates that Tarantool should be installed if application distribution doesn't contain tarantool binary; Tarantool version is got from VERSION file that is placed in distribution by Cartridge CLI;

• cartridge_create_user_group_for_tgz (boolean, default: false): flag indicates that specified user and group should be created;

• cartridge_app_user (string, default: tarantool): application user;

• cartridge_app_group (string, default: tarantool): application group;

• cartridge_data_dir (string, default: /var/lib/tarantool): directory where instances working directories are placed;

• cartridge_memtx_dir_parent (string): directory where instances memtx directories are placed;

• cartridge_vinyl_dir_parent (string): directory where instances vinyl directories are placed;

• cartridge_wal_dir_parent (string): directory where instances WAL directories are placed;

• cartridge_log_dir_parent (string): directory where instances logs are placed;

• cartridge_run_dir(string, default: /var/run/tarantool): directory where PID and socket files are stored;

• cartridge_conf_dir (string, default: /etc/tarantool/conf.d): path to instances configuration;

• cartridge_app_install_dir (string, default: /usr/share/tarantool): directory where application distributions are placed;

• cartridge_app_instances_dir (string, default: /usr/share/tarantool): directory where instances distributions are placed in case of multiversion approach;

• cartridge_configure_systemd_unit_files (boolean, default: true): flag indicates that systemd unit files should be configured;

• cartridge_systemd_dir (string, default: /etc/systemd/system): directory where systemd-unit files should be placed;

• cartridge_configure_tmpfiles (boolean, default: true): flag indicates that tmpfiles config should be configured for application run dir;

• cartridge_tmpfiles_dir (string, default: /usr/lib/tmpfiles.d/): a directory where tmpfile configuration should be placed;

• cartridge_keep_num_latest_dists (number, default: 2): the number of application versions that should be kept during distribution rotation.

Instances configuration

• cartridge_defaults (dict, default: {}): default configuration parameters values for instances;
• config (dict, required): instance configuration;
• zone (string): instance zone (available since Cartridge 2.4.0);
• cartridge_extra_env (dict): environment variables for instance service (should be set to an empty dictionary to remove set values);
• cartridge_runtime_params (dict): instance configuration parameters to patch in runtime;
• restarted (boolean): flag indicates if instance should be restarted or not (if this flag isn't specified, instance will be restarted if it's needed to apply configuration changes);
• expelled (boolean, default: false): a boolean flag that indicates if instance must be expelled from topology;
• disabled (boolean, default: false): a boolean flag that indicates if instance should be disabled;
• stateboard (boolean, default: false): a boolean flag that indicates that the instance is a stateboard;
• instance_start_retries (number, default: 10): retries to check that all instances become started;
• instance_start_delay (number, default: 5): delay before retry to check that all instances become started;
• [DEPRECATED] instance_start_timeout (number, default: 60): time in seconds to wait for instance to be started;
• cartridge_wait_buckets_discovery (boolean, default: true): a boolean flag that indicates if routers should wait for buckets discovery after vshard bootstrap;
• instance_discover_buckets_retries (number, default: 20): retries to check that instances discover buckets;
• instance_discover_buckets_delay (number, default: 5): delay before retry to check that instances discover buckets;
• [DEPRECATED] instance_discover_buckets_timeout (number, default: 60): time in seconds to wait for instance to discover buckets;
• allowed_members_states (list-of-strings, default: ['RolesConfigured']): list of allowed instance states;
• wait_members_alive_retries (number, default: 10): retries to check that all instances become alive;
• wait_members_alive_delay (number, default: 5): delay before retry to check that all instances become alive.

Replicasets configuration

• replicaset_alias (string): replicaset alias, will be displayed in Web UI;
• failover_priority (list-of-strings): failover priority;
• roles (list-of-strings, required if replicaset_alias specified): roles to be enabled on the replicaset;
• all_rw (boolean): indicates that that all servers in the replicaset should be read-write;
• vshard_group (string): name of vshard group the replicaset belongs to;
• weight (number): vshard replicaset weight (matters only if vshard-storage role is enabled);
• twophase_netbox_call_timeout (number): time in seconds to wait netbox call while two-phase commit (Cartridge 2.5+ is required);
• twophase_upload_config_timeout (number): time in seconds to wait config upload while two-phase commit (Cartridge 2.5+ is required);
• twophase_apply_config_timeout (number): time in seconds to wait config apply while two-phase commit (Cartridge 2.5+ is required);
• edit_topology_healthy_timeout (number, default: 60): time in seconds to wait until a cluster become healthy after editing topology;
• [DEPRECATED] edit_topology_timeout (number): the same timeout as edit_topology_healthy_timeout;
• edit_topology_allow_missed_instances (boolean, default: false): if set, then instances that aren't started yet are ignored on editing topology. A warning message is shown in case of error.

Cluster configuration

• connect_to_membership_retries (number, default: 3): retries to connect to membership;
• connect_to_membership_delay (number, default: 5): delay before retry to connect to membership;
• cartridge_bootstrap_vshard (boolean, default: false): a boolean flag that indicates if vshard should be bootstrapped;
• bootstrap_vshard_retries (number, default: 3): retries to bootstrap vshard;
• bootstrap_vshard_delay (number, default: 5): delay before retry to bootstrap vshard;
• cartridge_app_config (dict): application config sections to patch;
• cartridge_app_config_path (string): path to application config to upload;
• cartridge_app_config_upload_mode (string): mode of config uploading (lua, http or tdg);
• cartridge_app_config_upload_url (string): url of instance to upload config (http://127.0.0.1:{control_instance.http_port}/admin/config by default);
• cartridge_app_config_upload_http_timeout (number, default: 30): time in seconds to wait config upload in HTTP mode;
• cartridge_tdg_token (string): token to upload config by HTTP in TDG;
• cartridge_auth: (dict): authorization configuration;
• cartridge_failover_params (dict): failover parameters;
• [DEPRECATED] cartridge_failover (boolean): a boolean flag that indicates if eventual failover should be enabled or disabled;
• cartridge_failover_promote_params (dict): failover promote params;
• wait_cluster_has_no_issues_retries (number, default: 60): retries to check cluster has no issues;
• wait_cluster_has_no_issues_delay (number, default: 5): delay to retry cluster issues check;
• allow_warning_issues (boolean, default: false): allow issues with warning level;
• show_issues(boolean, default: true): log cluster issues as a warnings.

Eval step configuration

• cartridge_eval_file (string): path to file with Lua code to eval (isn't used if cartridge_eval_body is specified);
• cartridge_eval_body (string): code to eval;
• cartridge_eval_args (list, default: []): function arguments;
• cartridge_eval_with_retries (boolean, default: false): flag indicates that eval should be performed with retries;
• cartridge_eval_retries (number, default: 3): number of eval retries;
• cartridge_eval_delay (number, default: 5): eval retries delay.

Flags to ignore errors of checks for dangerous topology changes

• cartridge_force_advertise_uris_change (boolean, default: false): flag that disable check for advertise uris change;
• cartridge_ignore_extra_cluster_instances (boolean, default: false): flag that disable check for instances from the cluster that are not in inventory;
• cartridge_ignore_extra_cluster_replicasets (boolean, default: false): flag that disable check for replicasets from the cluster that are not in inventory;
• cartridge_ignore_renamed_replicasets (boolean, default: false): flag that disable check for replicasets that was renamed in cluster, but not renamed in inventory.

Steps configuration

• cartridge_delivered_package_path (string): remote path to file of delivered package (use it to run update_package step without deliver_package step);
• cartridge_control_instance (dict): some instance that can be used for editing topology and configuring cluster (use it to skip set_control_instance step).

Backups configuration

• cartridge_remote_backups_dir (string, default: /opt/tarantool/backups): directory to store backups on the remote;
• cartridge_fetch_backups (boolean, default: false): flag indicates that backups should be fetched on the local machine;
• cartridge_fetch_backups_dir (string, default: backups/): a directory on the local machine where backups should be fetched if cartridge_fetch_backups is true; this path is relative to the playbook path.
• cartridge_restore_backup_path (string): path to the instance backup archive on the remote machine;
• cartridge_restore_backup_path_local (string): path to the instance backup archive on the local machine;
• cartridge_force_restore (boolean, default: false): flag indicates that conflicting files should be overwritten;
• cartridge_allow_alien_backup (boolean, default: false): flag indicates that backup of instance with another name can be used;
• cartridge_skip_cleanup_on_restore (boolean, default: false): flag indicates that cleanup before restoring should be skipped;
• cartridge_paths_to_keep_before_restore (list-of-strings, default: []): list of paths that are absolute or relative to work/memtx/vinyl/wal directory that shouldn't be removed before instance restore (.tarantool.cookie will be kept independently of this variable); it's possible to use bash patterns, e.g. *.control.